elevenlabs.io

•

API Key

About ElevenLabs

ElevenLabs is an AI audio company offering high-quality text-to-speech synthesis, instant and professional voice cloning, multilingual dubbing, and conversational AI voice agents. It is used by content creators, publishers, game developers, and enterprises to produce realistic spoken audio at scale. The platform operates on a unified credit system where 1 character of text equals approximately 1 credit. Features span voice generation, sound effects, AI music, and agent-based voice interactions accessible via API.

AI Agent Use Cases

- Text-to-speech synthesis with 3,000+ AI voices in 32 languages

- Instant voice cloning from a short audio sample

- Professional voice cloning (Creator plan and above)

- Multilingual dubbing with synchronized lip-matching for video

- Conversational AI voice agents with real-time interaction

- API access for integrating voice into custom applications

Available Actions

These are the specific actions that AI agents can perform with this tool

Get Voice Information

2 inputs

Retrieve metadata information about a specific voice using its ID

Inputs

voice_id

ID of the voice to be retrieved. This parameter is required.

xi-api-key

API key for authentication. This parameter should be passed in the headers.

Get Voice Settings

1 input

Retrieve the settings for a specific voice.

Inputs

voice_id

Voice ID to be used to fetch the settings.

List Agents

3 inputs

Retrieve a list of agents along with their metadata

Inputs

cursor

Used for fetching the next page. Cursor is returned in the response.

page_size

How many Agents to return at maximum. Cannot exceed 100, defaults to 30.

Search by agent's name.

List Agents

3 inputs

Fetch a list of agents with their metadata

Inputs

cursor

Used for fetching the next page. The cursor is returned in the response.

page_size

The maximum number of Agents to return. Cannot exceed 100, defaults to 30.

Search by agent's name.

List Models

1 input

Retrieve a list of available models

Inputs

xi-api-key

The API key used for authentication

List Models

Get a list of all available models and their details

List Voices

10 inputs

Get a list of all available voices with options for search, filtering, and pagination.

Inputs

next_page_token

The next page token for pagination, received from the previous response.

page_size

Maximum number of voices to return. Defaults to 10, cannot exceed 100.

Search term to use for filtering voices by various attributes such as name or description.

sort

Field to sort by, either 'created_at_unix' or 'name'.

sort_direction

Direction for sorting, 'asc' or 'desc'.

voice_type

Voice type to filter by, options include 'personal', 'community', 'default', 'workspace', 'non-default'.

List Voices

10 inputs

Request to get a list of available voices, allowing search, filtering, and pagination.

Inputs

next_page_token

The next page token to use for pagination. Returned from the previous request.

page_size

How many voices to return at maximum. May include more voices on page 0 due to default voices.

Search term to filter voices by. Searches in name, description, labels, category.

sort

Field to sort by, not all fields available for older voices.

sort_direction

Direction to sort the voices in.

voice_type

Type of the voice to filter by.

Text to Speech Conversion

17 inputs

Convert provided text into speech using a specified voice.

Inputs

voice_id

ID of the voice to be used for the text to speech conversion.

text

The text that will get converted into speech.

enable_logging

When set to false, the request will be processed in zero retention mode which disables history features.

optimize_streaming_latency

Level of latency optimization to apply. Possible values range from 0 to 4 with increasing optimizations.

output_format

Output format of the generated audio. Example formats include mp3_44100_128 and mp3_22050_32.

model_id

Identifier of the model used for text to speech conversion.

language_code

ISO 639-1 language code to enforce a language for the model.

voice_settings

Settings to override stored voice settings for the request.

pronunciation_dictionary_locators

List of dictionaries to apply to the text for pronunciation.

seed

Seed for deterministic sampling, ensuring repeatability of requests with the same seed. Range: 0 - 4294967295.

previous_text

Text preceding the current request's text to improve speech continuity.

next_text

Text following the current request's text to improve speech continuity.

previous_request_ids

IDs of previous requests to enhance speech continuity. Maximum 3 IDs.

next_request_ids

IDs of following requests for speech continuity. Maximum 3 IDs.

apply_text_normalization

Controls text normalization. Allowed values: 'auto', 'on', 'off'.

apply_language_text_normalization

Controls language text normalization for proper pronunciation in some languages.

use_pvc_as_ivc

Use IVC version in place of PVC version of the voice.

Voice Conversion

11 inputs

Transform audio from one voice to another with control over emotion, timing, and delivery.

Inputs

voice_id

ID of the voice to be used.

xi_api_key

API key for authentication, included in the request header.

enable_logging

When set to false, zero retention mode is used, preventing history features for the request.

optimize_streaming_latency

Turn on latency optimizations. Possible values: 0-4. Defaults to None.

output_format

Output format of the generated audio.

audiofile

The audio file with the content and emotion to control the generated speech.

model_id

Identifier of the model that will be used, supporting speech to speech.

voice_settings

Voice settings overriding stored settings for the given voice, sent as a JSON encoded string.

seed

For deterministic sampling. Must be between 0 and 4294967295.

remove_background_noise

If set, will remove the background noise from your audio input.

file_format

Format of input audio. Options: ‘pcm_s16le_16’ or ‘other’.